RhoB regulates uPAR signalling

Daniela Alfano1,2,*, Pia Ragno2, M. Patrizia Stoppelli3 and Anne J. Ridley1,`

1Randall Division of Cell and Molecular Biophysics, King’s College London, New Hunt’s House, Guy’s Campus, London SE1 1UL, UK2Department of Chemistry and Biology, University of Salerno, Via Ponte don Melillo 1, 84084 Fisciano, Salerno, Italy3Institute of Genetics and Biophysics Adriano Buzzati-Traverso, Consiglio Nazionale delle Ricerche (CNR), Via P. Castellino 111, 80131 Naples, Italy

*Present address: Institute of Genetics and Biophysics Adriano Buzzati-Traverso, Consiglio Nazionale delle Ricerche (CNR), Via P. Castellino 111, 80131 Naples, Italy`Author for correspondence (anne.ridley@kcl.ac.uk)

Accepted 12 January 2012Journal of Cell Science 125, 2369–2380� 2012. Published by The Company of Biologists Ltddoi: 10.1242/jcs.091579

SummaryUrokinase-type plasminogen activator (uPA) and its receptor, uPAR, play important roles in promoting cancer cell adhesion, migration andinvasion. Rho GTPases are key coordinators of these processes; the Rho GTPase Rac1 has previously been implicated in uPA- and/or uPAR-induced migratory or morphological cell responses. We used RNAi to deplete 12 different Rho GTPases to screen for effects on uPA-stimulated migration, and found that depletion of RhoB significantly reduces uPA-induced migration and invasion of prostate carcinoma

cells. RhoB depletion did not affect the expression or surface levels of uPAR but reduced the uPAR-induced increase in levels of severalintegrins and inhibited uPAR signalling to the actin regulator cofilin, the cell-adhesion signal-transduction adaptor molecule paxillin and theserine/threonine kinase Akt. uPAR rapidly activated RhoB and increased RhoB expression. RhoB depletion also reduced cell adhesion to and

spreading on vitronectin, which is a uPAR ligand. This correlated with decreased association between integrins and uPAR and reducedintegrin b1 activity. Our results indicate that RhoB is a key regulator of uPAR signalling in cell adhesion, migration and invasion.

Key words: Rho GTPases, Cell migration, Signal transduction, uPA, uPAR

IntroductionUrokinase-type plasminogen activator (uPA) stimulates changes tocells and their environments both through its proteolytic activity

and through binding to its cell surface receptor, the uPA receptor(uPAR). uPA is a protease that converts the pro-enzymeplasminogen into plasmin, a wide-spectrum serine protease able

to degrade most of the extracellular matrix (ECM) components andactivate latent collagenases. uPA and plasmin are involved inthrombolysis, inflammation, cell migration, tissue remodelling,cancer invasion and vascularization (Irigoyen et al., 1999). uPA is

secreted in the pro-enzyme form (pro-uPA), which can beactivated in the extracellular milieu by a single proteolyticcleavage occurring between Lys158 and Ile159, thus generating

a two-chain enzyme. uPA is a multi-domain protein that includesan amino-terminal growth factor-like domain (GFD; residues 1–49) followed by a ‘kringle’ region (residues 50–131) linked by the

‘connecting peptide’ (residues 135–158) to the catalytic proteasedomain (residues 158–411). High affinity binding to uPAR occursthrough the GFD of uPA and does not involve the catalytic domainor alter its catalytic activity (Stoppelli et al., 1985; Vassalli et al.,

1985; Vincenza Carriero and Stoppelli, 2011). uPAR is a three-domain (D1, D2 and D3) glycosylphosphatidylinositol-anchoredprotein that binds uPA or the uPA amino-terminal fragment (ATF;

residues 1–135), mainly through its N-terminal D1 domain (Llinaset al., 2005; Vincenza Carriero et al., 2009). Independently of itscatalytic domain, uPA acts through uPAR to induce dynamic

reorganization of the actin cytoskeleton, cell adhesion to theextracellular matrix (ECM) and cell migration. uPAR also bindsthe ECM protein vitronectin (VN), thus mediating cell adhesion

(Montuori et al., 2005). uPAR activates a number of signallingmolecules, including Rho family GTPases, Src family kinases andERK1/2 Ser/Thr kinases (Blasi and Sidenius, 2010; Smith and

Marshall, 2010). Because uPAR is not a transmembrane receptor,uPA- and VN-dependent signalling require physical and/orfunctional interaction of uPAR with a variety of transmembrane

receptors, such as integrins, the EGF receptor and fMLP receptors(Alfano et al., 2005; Franco et al., 2006; Montuori et al., 2011).

In human cancer, increased expression of uPA and uPAR isassociated with a high risk of metastases and an unfavourableclinical outcome (Harbeck et al., 2002; Danø et al., 2005).

Prostate cancer continues to be one of the deadliest cancers andcurrently ranks as the second most frequent cause of cancer-related death in males in the United States. The uPA–uPAR

system could serve as a useful molecular marker to predict a poorprognosis for prostate cancer and might be a therapeutic target forprostate cancer intervention (Sheng, 2001; Sehgal et al., 2006;Almasi et al., 2011; Rabbani et al., 2010). uPAR and uPA

overexpression seems to be associated with a more aggressivephenotype of prostate cancer cells, because they stimulate cellinvasion, survival and tumorigenicity (Pulukuri et al., 2005).

Rho family GTPases are crucial regulators of cytoskeletal andadhesion dynamics, thereby stimulating cell migration and

invasion (Heasman and Ridley, 2008). Although there are atleast 20 human Rho GTPases, only a few have been rigorouslytested for their involvement in migratory responses (Vega and

Ridley, 2008). The uPA–uPAR system has been reported to actthrough Rac, Cdc42 and RhoA (Smith et al., 2008; Smith andMarshall, 2010). uPAR overexpression by itself induces an

increase in Rac-dependent lamellipodia and cell migration, andincreases the level of activated Rac (Kjoller and Hall, 2001;Aguirre-Ghiso et al., 2003). Conversely, downregulation of uPAR

by RNA interference (RNAi) results in the loss of actin-richmembrane ruffles, and reduces Rac activity (Smith et al., 2008).RhoA, however, stimulates uPAR transcription whereas Rac1 does

Research Article 2369

Journ

alof

Cell

Scie

nce

not (Muller et al., 2000). To date, only RhoA and Rac1 have been

tested in uPA-induced migration (Kiian et al., 2003), whereas the

involvement of other Rho GTPases has not been investigated.

The aim of this work was to identify which Rho GTPase(s)

mediate uPA–uPAR-induced migration and invasion. Using

RNAi-mediated knockdown of 12 different Rho GTPases, we

identified RhoB as a key regulator of uPAR-dependent responses

in prostate cancer cells, including cell migration, invasion and

adhesion to vitronectin. RhoB affects uPAR-induced integrin

expression, association of uPAR with integrins and integrin

activity. In addition, uPAR activates RhoB and stimulates RhoB

expression. These results indicate a previously unknown role for

RhoB in uPAR responses.

ResultsuPA and uPAR expression regulates cell motility and

invasiveness of PC3 prostate cancer cells

One of the key functions of the uPA–uPAR system is to promote

invasion, a crucial process for tumour metastasis (Blasi and

Carmeliet, 2002). We investigated the contribution of uPA and

uPAR to the migration and invasion of the highly invasive human

prostate cancer cell line, PC3 (Dedhar et al., 1993). We first

assessed uPA and uPAR expression in PC3 cells. uPA and uPAR

were both expressed in PC3 cells and their expression was reduced

by specific short interfering RNAs (siRNAs) (Fig. 1A,B). We then

investigated the effects of uPA and uPAR depletion in invasion

assays. PC3 cells efficiently invaded through Matrigel towards

serum acting as a chemoattractant; this invasion was reduced by

50% following uPA depletion and 40% following uPAR depletion,

compared with control cells (Fig. 1C). To determine whether uPA

also affected cell migration in two dimensions we performed a

scratch-wound-healing assay with PC3 cells and monitored them

by time-lapse microscopy. uPA depletion substantially reduced

PC3 cell migration into the wound area, both in 10% FCS

(Fig. 1D) and 0.5% FCS (data not shown). uPA and uPAR are

therefore required for efficient PC3 cell migration and invasion.

uPAR-induced migration and invasion requires a specific

subset of Rho GTPases

The uPA–uPAR system is known to act through Rho GTPases to

affect cell migration and invasion (Smith and Marshall, 2010), but a

comprehensive analysis of Rho GTPase involvement downstream

Fig. 1. Expression of uPA and uPAR regulates migration

and invasion of PC3 cells. (A) To knockdown endogenous

uPA and uPAR, PC3 cells were transfected with a control

siRNA (siControl), a siRNA targeting uPA (si-uPA) or two

different siRNAs targeting uPAR (si-uPAR1 and 2). After

72 hours, cell lysates were immunoblotted with anti-uPA,

monoclonal R4 anti-uPAR or anti-tubulin and anti-GAPDH

antibodies as loading controls. (B) To determine the level of

secreted uPA following uPA knockdown, medium was

collected from cells 48 and 72 hours after siRNA transfection,

concentrated and analysed by immunoblotting. Relative uPA

levels were quantified by densitometric scanning (numbers

below the blot, normalized at each time point to siControl).

(C) PC3 cells transfected with control, uPA or uPAR siRNAs

were plated on Matrigel-coated Transwell filters containing

10% FCS in the bottom well. (Left) The invasion values are the

means 6 s.e.m. of three experiments performed in triplicate,

normalized to the number of invading cells in the absence of

FCS, shown as fold change relative to control. **P#0.01

compared with siControl-transfected cells. (Right)

Representative images of cells on the Transwell filters. Scale

bars: 50 mm. (D) A confluent monolayer of PC3 cells was

wounded using a pipette tip and cells were monitored by time-

lapse microscopy, acquiring an image every 3 minutes for

15 hours. (Left) Representative images 15 hours after scratch

wounding. (Right) Scratch wound areas were determined from

phase-contrast images taken at the indicated time points from

movies, and are shown as a percentage of the initial

wound area.

Journal of Cell Science 125 (10)2370

Journ

alof

Cell

Scie

nce

of uPAR has not been carried out. PC3 cells were therefore

transiently transfected with siRNAs targeting 12 of the 20 known

Rho GTPase family members, in order to assess the involvement of

specific Rho GTPases in uPA-stimulated migration. These 12 Rho

GTPases were chosen on the basis of phenotypes observed in other

experiments knocking down Rho GTPases in PC3 cells in our

laboratory (Vega and Ridley, 2011; Bai et al., 2011) (unpublished

data). The efficiency of Rho GTPase knockdown was determined

by western blotting of all Rho GTPases except for RhoD, for which

no reliable antibody was available (Fig. 2A). Specific siRNAs

strongly reduced the expression of the Rho GTPases analysed. To

determine which Rho GTPases affect uPAR-dependent cell

migration, PC3 cells were plated on collagen IV-coated

Transwells and allowed to migrate toward the N-terminal region

of uPA, known as the amino-terminal fragment (ATF, residues 1–

135). ATF binds uPAR (Stoppelli et al., 1985; Vassalli et al., 1985)

but lacks the catalytic uPA domain, thus its pro-migratory effect is

only due to its interaction with uPAR and not to its proteolytic

activity. ATF-stimulated migration of PC3 cells; Rac1, Cdc42,

RhoB, RhoC or RhoF depletion inhibited ATF-induced migration.

By contrast, depletion of RhoA, RhoD, RhoE, RhoG, RhoJ, RhoQ

or RhoU did not significantly (P.0.05) affect migration towards

ATF (Fig. 2B; siRNAs are listed in supplementary material Table

S1). Rac1 and Cdc42 are known to affect uPAR-dependent

migration (Kjoller and Hall, 2001; Aguirre-Ghiso et al., 2003;

Chandrasekar et al., 2003; Kiian et al., 2003), thus we further

analysed the functions of RhoB, RhoC and RhoF. We found that

RhoB, RhoC and RhoF depletion also reduced invasion through

Matrigel in an ATF gradient, whereas RhoA did not (Fig. 2C).

uPAR-stimulated migration and invasion therefore requires a

subset of specific Rho GTPases; we chose to focus our study on

the involvement of RhoB in uPA responses, because RhoB

depletion seems to impair uPAR-dependent invasion more

strongly than RhoC and RhoF, suggesting a specific role in

uPAR signalling. Expression of GFP–RhoB rescued the

chemotactic response of RhoB-depleted cells to ATF, and thus

knockdown of RhoB and not an off-target gene is responsible for

the observed effect (Fig. 2D). Western blot analysis showed GFP–

RhoB expression in control and RhoB-knockdown cells (Fig. 2D,

right panel).

RhoB contributes to uPAR signalling

To investigate how RhoB affected uPAR-induced migration and

invasion, we tested the effects of knocking down RhoB on uPAR

Fig. 2. uPAR-induced migration and invasion requires a specific

subset of Rho GTPases. PC3 cells (36105) were seeded in six-well

plates and transfected with siRNAs targeting the indicated Rho

GTPases or with a control siRNA (siControl). (A) Transfected cells

were lysed and analysed 72 hours after transfection by immunoblotting

for the indicated Rho GTPases, or tubulin as a loading control. Results

shown are for siRNA-1 for each Rho GTPase (supplementary material

Table S1). Similar results were obtained with other siRNAs to each

gene. (B) Migration assay. Transfected cells were plated on collagen-

coated Transwell filters with and without ATF in the bottom well. The

values are the means 6 s.e.m. of three experiments performed in

triplicate. Results with the different siRNAs for each Rho GTPase

listed in supplementary material Table S1 were pooled (see

supplementary material Fig. S1 for results with individual siRNAs) and

are shown as a percentage of siControl without ATF; #P#0.05

compared with siControl 2ATF; *P#0.05 compared with siControl

+ATF. (C) Transfected cells were plated on Matrigel-coated Transwell

filters to assess their invasion ability in an ATF gradient. Both

migration and invasion results are pooled from experiments using all of

the siRNAs to each gene listed in supplementary material Table S1.

(D) PC3 cells were transfected with RhoB-targeting siRNAs (siRhoB1

or siRhoB2) or control siRNA (siControl) and after 24 hours cells were

transfected with a construct encoding GFP–RhoB or GFP alone; 24

hours later, cells were allowed to migrate towards ATF to examine

their chemotaxis (left). The levels of endogenous RhoB and ectopic

GFP–RhoB protein were determined by immunoblotting (right). In C

and D 100% values represent random cell migration or invasion for

each siRNA treatment in the absence of ATF. The values are the means

6 s.e.m. of three experiments performed in triplicate; *P#0.05,

**P#0.01, ***P#0.001 compared with siControl.

RhoB regulates uPAR signalling 2371

Journ

alof

Cell

Scie

nce

levels and uPAR-induced signalling. RhoB depletion did not

affect uPAR protein expression as shown by western blot

(Fig. 3A) or affect surface levels of uPAR (Fig. 3B). It also did

not alter mRNA levels as determined by real-time quantitative

PCR (qPCR) analysis (Fig. 3C).

uPAR signalling is known to involve integrins (Smith and

Marshall, 2010). Because RhoB depletion impaired uPAR-

mediated responses but did not affect uPAR levels, we

investigated the possibility that RhoB could regulate integrin

levels. ATF treatment increased the cell surface levels of b1, b5

and av integrins as detected by flow cytometry (Fig. 3D). This in

part reflected changes in total levels of these integrins (Fig. 3E).

RhoB depletion reduced the ATF-induced increase in the surface

levels of b1, b5 and av integrins (Fig. 3D). This reduction in

integrin levels could explain how RhoB specifically affects

uPAR-dependent migration and invasion but not basal migration.

We then investigated the involvement of two regulators of cell

migration and adhesion, cofilin and paxillin, in uPAR-dependent

signalling. Cofilin severs actin filaments to increase sites for

actin polymerization in lamellipodia, and it is inhibited by

phosphorylation on Ser3 (Bamburg, 1999). Paxillin localizes to

integrin-mediated adhesion sites and is a multi-domain adaptor

and scaffolding molecule that is regulated by tyrosine

phosphorylation (Schaller and Parsons, 1995). Western blot

analysis of ATF-stimulated PC3 cells showed changes in the

phosphorylation of both these proteins. ATF transiently

stimulated cofilin Ser3 phosphorylation, returning to basal

levels by 60 minutes, and induced an increase in paxillin

phosphorylation by 1 minute, which was still maintained at

60 minutes (Fig. 4A).

uPA has previously been shown to activate the

phosphoinositide 3-kinase (PI3K)–Akt pathway in other cell

types (Alfano et al., 2006). Accordingly, we found that ATF

stimulated Akt phosphorylation in PC3 cells (Fig. 4A). RhoB

depletion prevented ATF-induced cofilin, paxillin and Akt

phosphorylation (Fig. 4B,C). These results indicate that RhoB

contributes to three different aspects of uPAR signalling: actin

dynamics (cofilin), cell adhesion (paxillin) and cell survival

(Akt).

ATF increases RhoB activation and expression

Because RhoB is required for uPAR responses, we assessed

RhoB activation following ATF stimulation. RhoB activation

was observed in PC3 cells between 1 minute (data not shown)

and 3 minutes (Fig. 5A) after ATF stimulation with the same

concentration (10 nM) used to stimulate cell migration; 1 nM

ATF did not stimulate RhoB (data not shown). RhoB has

previously been shown to be upregulated at the mRNA and

protein levels in response to several growth factors (Wheeler and

Ridley, 2004), thus we investigated whether ATF also affected

RhoB expression. qPCR analysis showed that ATF increased

RHOB mRNA levels in PC3 cells as early as 1 hour (threefold),

with maximal levels at 6 hours (sevenfold; Fig. 5B); a parallel

increase of RhoB protein, peaking at 6 hours after stimulation

was also observed (Fig. 5C). The early induction of RHOB

mRNA at 1 hour is consistent with it being an immediate early

Fig. 3. RhoB knockdown alters integrin levels but not

uPAR expression. PC3 cells were transfected with control

siRNA (siControl) or three RhoB-targeting siRNAs (siRhoB1,

siRhoB2 or siRhoB3) and analysed after 48 hours.

(A) Transfected cell lysates were analysed by immunoblotting

with polyclonal anti-RhoB antibody, monoclonal R4 anti-

uPAR antibodies, or anti-tubulin antibodies as a loading control

(left panels). Relative uPAR and tubulin levels were quantified

by densitometric scanning and uPAR values were normalized

to tubulin (right panel). Values are the ratio of uPAR levels in

siRHOB-treated cells to those in siControl cells (means 6

s.e.m. of three separate experiments). (B) Cell surface levels of

uPAR were determined by flow cytometry. Values are the

mean fluorescence of the population 6 s.d. from two separate

experiments each performed in duplicate. Results were

normalised to siControl values. (C) uPAR mRNA levels were

quantified by qPCR and normalized to the corresponding

GAPDH mRNA levels. Values are the ratio of uPAR mRNA

levels in siRHOB-treated cells to siControl cells (means 6

s.e.m. of three separate experiments). (D) PC3 cells were

serum-starved for 24 hours, then stimulated with 10 nM ATF

or left unstimulated for 9 hours. Cells were stained for surface

levels of b1, b5 and av integrins. Values are the mean

fluorescence of the population 6 s.d. of two separate

experiments each performed in duplicate, relative to control

cells not treated with ATF. (E) Cell lysates from control and

ATF-stimulated cells were analysed by immunoblotting with

antibodies to the indicated integrins. Tubulin was used as a

loading control.

Journal of Cell Science 125 (10)2372

Journ

alof

Cell

Scie

nce

gene (Huang and Prendergast, 2006; Vasilaki et al., 2010). The

later upregulation at 6 hours could reflect another layer of

transcriptional regulation. uPA has previously been reported to

stimulate RhoA activation in MCF-7 breast cancer cells (Jo et al.,

2002), and Cdc42 and Rac activation in MDA-MB-231 breast

cancer cells (Sturge et al., 2002). We demonstrate here that ATF

both stimulates RhoB activation and increases RhoB expression.

uPAR expression regulates PC3 cell adhesion to

vitronectin

uPAR mediates adhesion of a variety of cell types to VN, because

it is a non-integrin VN receptor (Deng et al., 1996; Wei et al.,

1996; Madsen and Sidenius, 2008; Blasi and Sidenius, 2010).

Consistent with this, uPAR overexpression in PC3 cells increased

cell elongation and spread area (Fig. 6A), and increased their

adhesion to VN (Fig. 6B). uPAR expression also increased cell

migration speed (Fig. 6C) and induced the extension of numerous

microspikes on the plasma membrane of PC3 cells plated on VN

(Fig. 6D). Endogenous uPAR contributes to PC3 cell adhesion to

VN, because the function-blocking 399R anti-uPAR antibody

inhibited adhesion of both PC3 and uPAR-transfected PC3 cells

to VN (Fig. 6E).

RhoB specifically regulates PC3 cell adhesion to

vitronectin

RhoB depletion has previously been reported to reduce the

adhesion of macrophages to ICAM-1- and serum-coated surfaces

(Wheeler and Ridley, 2007). Depletion of RhoB inhibited

adhesion of both PC3 and uPAR-overexpressing PC3 cells to

VN (Fig. 7A), suggesting that RhoB affects uPAR-mediated

adhesion. The function-blocking 399R anti-uPAR antibody did

not further reduce the adhesion of RhoB-depleted cells,

suggesting that the effect of RhoB on adhesion is

predominantly through uPAR (Fig. 7B). Consistent with this,

Fig. 4. RhoB depletion inhibits uPAR-induced

paxillin, Akt and cofilin phosphorylation.

(A) ATF stimulates cofilin, paxillin and Akt

phosphorylation in PC3 cells. Cells were serum-

starved for 24 hours, stimulated with 10 nM ATF

and lysed at the indicated time points. Cell lysates

were analysed by immunoblotting with antibodies

to the indicated phosphorylated proteins. GAPDH

was used as a loading control. Relative protein

levels were quantified by densitometric scanning.

Phosphorylated cofilin (p-cofilin) levels were

normalized to cofilin levels, phosphorylated

paxillin levels to paxillin levels and phosphorylated

Akt levels to Akt levels. Values are the ratio of

level of the phosphorylated proteins in ATF-

stimulated to unstimulated cells. (B) Cells were

transfected with control siRNA (siControl) or the

indicated siRNAs targeting RhoB. (Left)

Transfected cell lysates were analysed by

immunoblotting with polyclonal anti-RhoB

antibody. (Right and below) After 72 hours, cells

were serum-starved, stimulated with 10 nM ATF

and lysed at 1, 3 or 20 minutes, as indicated. Cell

lysates were analysed by immunoblotting with

antibodies to the indicated phosphorylated proteins.

Levels of total paxillin or cofilin were used as

loading controls. Relative protein levels were

quantified by densitometric scanning.

Phosphorylated cofilin, paxillin and Akt levels were

normalised respectively to total cofilin, paxillin and

Akt levels. Values are levels of the phosphorylated

proteins relative to those in unstimulated siControl

cells. Blots shown are representative of at least

three independent experiments. (C) Graph showing

densitometric analysis of immunoblots. Values are

the fold increase over basal levels (2ATF) for

phosphorylated cofilin, paxillin and Akt,

normalized to total protein levels of the respective

protein (n53 for each protein; means 6 s.d.

*P,0.05 versus control).

RhoB regulates uPAR signalling 2373

Journ

alof

Cell

Scie

nce

RhoB knockdown specifically decreased PC3 cell adhesion to

VN but not to collagen IV or uncoated plastic (Fig. 7C). RhoB-

depleted PC3 cells had a smaller spread area and were rounder

than control cells on VN (Fig. 7D). GFP–RhoB expression

rescued the spread area of RhoB-depleted cells (Fig. 7E),

showing that the effects of siRNA targeting of RhoB were

specific and not due to off-target effects. RhoB-depleted PC3

cells migrated more slowly than control cells (Fig. 7F), and again

this was rescued by GFP–RhoB. The smaller spread area of

RhoB-depleted cells was not due to a smaller cell volume or

diameter, as measured in suspension (Fig. 7G). However, RhoB

overexpression was unable to rescue the spread area of uPAR-

depleted cells (Fig. 7H), confirming that RhoB-mediated

adhesion of PC3 cells on VN is uPAR dependent. Hence,

RhoB depletion impaired PC3 cell adhesion specifically to VN,

which is a high affinity ligand of uPAR (Blasi and Sidenius,

2010). We propose that RhoB depletion reduces adhesion to VN,

thereby reducing cell migration speed.

RhoB depletion reduces the association of uPAR withintegrins

In order to elucidate the mechanism through which RhoB affectsuPAR-dependent adhesion and spreading, we investigated thepossibility that RhoB could affect the association of uPAR with

integrins, because integrins generally mediate uPAR signalling(Smith and Marshall, 2010). The level of uPAR in b1 integrin orav integrin immunoprecipitates was reduced in RhoB-depleted

cells compared with control cells (Fig. 8A). Interestingly, RhoBdepletion also reduced the association of talin with b1 and avintegrins (Fig. 8A). Given that talin is a crucial integrin-

activating protein (Calderwood, 2004) we investigated theactivation status of surface b1 integrin by flow cytometry.RhoB depletion reduced the level of active b1 integrin (Fig. 8B).Taken together these results suggest that RhoB affects uPAR-

dependent adhesion and spreading by reducing the interactionbetween uPAR and active integrins.

DiscussionuPA and uPAR are well known to stimulate the migration

and invasion of cancer cells thereby contributing to cancerprogression and metastasis, and to signal through integrins (Blasiand Carmeliet, 2002). Rac1 is required for uPAR-mediated

migratory responses (Smith and Marshall, 2010), but theinvolvement of Rho GTPases other than Rac1, Cdc42 andRhoA in uPA–uPAR signalling has not been tested. Weinvestigated the role of 12 different Rho GTPases in uPA–

uPAR responses in prostate carcinoma cells and found thatsilencing RhoB impaired uPAR-dependent signalling in cellmigration and invasion. uPAR stimulation induced RhoB

activation and also increased RhoB levels. We demonstrate thatRhoB mediates uPAR-induced upregulation of surface integrinlevels and signalling to paxillin, cofilin and Akt. It also controls

the association of uPAR and the integrin-activating protein talinwith integrins and modulates b1 integrin activation. These resultsdemonstrate that RhoB is a crucial mediator of uPAR signallingand that it acts through integrins.

uPAR can bind independently or simultaneously to its twoligands, uPA and VN (Madsen et al., 2007; Smith and Marshall,

2010). We demonstrate that RhoB contributes to both uPA-drivenuPAR responses and adhesion to VN. First, by using the amino-terminal fragment (ATF) of uPA, which binds to uPAR andactivates uPAR signalling but is devoid of proteolytic activity, we

could dissociate proteolytic effects of uPA from its effects onuPAR (Vincenza Carriero et al., 2009). This allowed us todemonstrate that uPAR-induced migration and invasion of PC3

cells required RhoB. Depletion of uPAR also inhibited uPA-induced migration and invasion, demonstrating the importance ofuPAR signalling in PC3 cell migratory responses.

We show that RhoB is required for adhesion of PC3 cells toVN but not to other substrates, and for adhesion to VN of uPAR-overexpressing cells. However, RhoB overexpression does not

rescue adhesion to VN after uPAR knockdown. Because uPAR isknown to contribute to VN adhesion (Madsen et al., 2007) theseresults suggest that RhoB is specifically involved in mediating

uPAR-dependent PC3 cell adhesion to VN. RhoB has previouslybeen implicated in regulating cell adhesion in other cell types.For example, macrophages and fibroblasts from RhoB-null mice

have defective adhesion and spreading (Liu et al., 2001; Wheelerand Ridley, 2007), and RhoB depletion inhibits adhesion ofhuman lung epithelial cells (Bousquet et al., 2009). In

Fig. 5. uPAR increases RhoB activation and expression. PC3 cells were

serum-starved for 24 hours, then stimulated with 10 nM ATF and harvested at

the indicated time points. (A) RhoB activity in ATF-stimulated or untreated

cells was evaluated by GST-RBD pulldown assay. The levels of total RhoB or

active RhoB pulled down by GST-RBD were analysed by immunoblotting

with anti-RhoB antibodies. Tubulin was used as a loading control. (B) RHOB

mRNA levels in ATF-treated cells were analysed by qPCR. RHOB mRNA

levels were normalized to the corresponding GAPDH mRNA levels. Values

are shown as fold increase relative to RHOB mRNA levels in untreated cells.

(C) RhoB protein levels in ATF-treated PC3 cells were analysed by

immunoblotting with RhoB-specific antibodies. Tubulin was used as a

loading control. Data are representative of three independent experiments.

Journal of Cell Science 125 (10)2374

Journ

alof

Cell

Scie

nce

macrophages, RhoB has been postulated to affect adhesion by

affecting surface levels of b-integrins (Wheeler and Ridley,

2007). We demonstrate that RhoB mediates ATF-induced

upregulation of surface integrin levels, explaining how RhoB

specifically affects uPAR responses, and hence uPAR-dependent

signalling. We also found that RhoB depletion reduces the level

of activated b1 integrin; moreover the association of uPAR, as

well as the integrin-activating protein talin, with integrins is

strongly decreased in RhoB-depleted cells. RhoB could modulate

uPAR–integrin and talin–integrin association through its effects

on surface integrin levels and activity.

Integrins traffic dynamically between the plasma membrane

and endosomes, and this trafficking is regulated by multiple

stimuli (Caswell et al., 2009). RhoB might affect integrin surface

levels and activity by regulating integrin endosomal trafficking.

RhoB is known to regulate trafficking of tyrosine kinase

receptors, including EGFR and PDGFRb, to late endosomes

and/or lysosomes and hence affect receptor signalling (Wherlock

et al., 2004; Huang et al., 2007). The non-receptor tyrosine kinase

Src is also dependent on RhoB for trafficking (Sandilands et al.,

2004). RhoB has been postulated to affect protein trafficking by

regulating actin polymerization on endosomes (Sandilands et al.,

2004; Ridley, 2006).

We observed a striking ability of uPAR to activate RhoB:

RhoB was rapidly activated by the addition of ATF. RhoB

activation is also rapidly stimulated by exposure to ultra violet

(UVB) and ionizing radiation, which, similar to ATF,

subsequently increase RhoB protein levels (Canguilhem et al.,

2005; Monferran et al., 2008). The activation of RhoB by

uPAR could be mediated by integrins as co-receptors for uPAR.

uPAR–integrin-b1 interaction has been reported to be required

for uPA-induced activation of Rac1 and Cdc42 in MDA-MB-231

breast cancer cells (Sturge et al., 2002), and uPAR association

with integrin b2 is involved in Rac1 and Cdc42 activation in

microvascular endothelial cells (Margheri et al., 2006);

moreover, uPAR-induced Rac activation is dependent on

integrin b3 in a variety of cancer cell lines (Smith et al., 2008).

As well as activating RhoB, ATF induced RhoB mRNA and

protein expression. RhoB gene expression is well known to be

induced rapidly by a variety of different stimuli, including growth

factors (Jahner and Hunter, 1991), stress stimuli such as UV

irradiation and TGFb (Jahner and Hunter, 1991; Fritz et al., 1995;

Vasilaki et al., 2010). RhoB mRNA and protein have short half-

lives (Fritz et al., 1995; Lebowitz et al., 1995) and thus it could

be important to increase RhoB expression to maintain the levels

required for sustained cell migration. Whether the uPAR-induced

increase in RhoB levels is indeed required for uPAR-stimulated

cell migration remains to be established, but interestingly RhoB

expression is also induced by TGFb, and RhoB contributes to

TGFb-induced migration of HaCaT keratinocytes and DU145

prostate cancer cells (Vasilaki et al., 2010).

RhoB is often considered a tumour suppressor because its

expression is downregulated in some cancers, and this has been

linked to its role in stimulating apoptosis (Huang and

Prendergast, 2006; Vega and Ridley, 2008; Bousquet et al.,

2009). By contrast, RhoB can either stimulate or inhibit cancer

migration or invasion depending on the cellular context. For

example, overexpression of RhoB suppresses invasion and/or

migration of gastric cancer cell lines (Zhou et al., 2011) and

hepatocellular carcinoma cell lines (Connolly et al., 2010). By

Fig. 6. uPAR expression regulates PC3 cell adhesion to

vitronectin. (A) Representative fluorescence images of PC3

cells expressing uPAR–GFP or GFP alone as a control, plated

on VN-coated plates. (B) PC3 cells expressing uPAR–GFP or

GFP alone were plated onto VN-coated wells. After 1 hour,

attached cells were fixed and GFP-expressing cells were

counted. Data are from two different experiments, each carried

out in triplicate, analysing 15 cells per field (,90 cells in total).

Values are means 6 s.d.; *P,0.05 compared with control.

(C) Cell speed was determined by time-lapse microscopy,

acquiring an image every 10 minutes over a period of 16 hours.

Data are from 30 cells in two independent experiments. Values

are means 6 s.e.m., **P,0.01. (D) PC3 cells were transfected

with a construct encoding uPAR–GFP or the GFP vector; 24

hours after transfection, cells were plated on VN-coated

coverslips in the absence of serum. After a further 24 hours,

cells were fixed and stained for F-actin. Scale bars: 20 mm.

(E) PC3 cells expressing uPAR–GFP or GFP alone were pre-

incubated for 30 minutes at 37 C with a polyclonal 399 anti-

uPAR antibody (25 mg/ml) or buffer as control, and then plated

onto VN-coated wells. After 1 hour, attached cells were fixed

and GFP-expressing cells were counted. Data are from two

different experiments, each carried out in triplicate, analysing

15 cells per field (,90 cells in total). Values are means 6 s.d.;

*P,0.05 compared with control or #P,0.05 compared with the

absence of anti-uPAR antibody.

RhoB regulates uPAR signalling 2375

Journ

alof

Cell

Scie

nce

Fig. 7. RhoB specifically regulates PC3 cell adhesion to vitronectin. (A) PC3 cells were co-transfected with RhoB-targeting siRNA (siRhoB1 or siRhoB2) or

control siRNA (siControl) and after 24 hours cells were transfected with a construct encoding uPAR–GFP or GFP alone; 24 hours later, cells were allowed to

adhere to VN-coated wells for 30 minutes. The attached cells were fixed and stained with Crystal Violet. The stain was eluted, and the absorbance at 595 nm was

measured by spectrophotometry. Values are the means 6 s.d. of three experiments performed in triplicate, *P,0.05, **P,0.01, ***P,0.001; #P,0.05,###P,0.001 compared with siControl. (B) PC3 cells were transfected with RhoB-targeting siRNA (siRhoB1 or siRhoB2) or control siRNA (siControl). After

48 hours they were incubated for 30 minutes at 37 C with a polyclonal 399 anti-uPAR antibody (25 mg/ml) or buffer as control, and then cell were plated onto

VN-coated wells and allowed to adhere for 1 hour. (C) Adhesion of PC3 cells transfected with control siRNA (siControl) or with RhoB-targeting siRNAs

(siRhoB2 or siRhoB3) was determined 48 hours after transfection by allowing cells to adhere for 1 hour to plates coated with different substrates, as indicated.

Values are the means 6 s.d. of three independent experiments, each performed in triplicate, *P,0.05, **P,0.01 compared with control. (D) Phase-contrast

images of PC3 cells transfected with control siRNA (siControl) or with RhoB-targeting siRNA (siRhoB1). 48 hours after transfection cells were seeded on VN-

coated plates and cultured without serum, and 6 hours later they were imaged by time-lapse microscopy for 24 hours (left). Scale bars: 50 mm. Cell area was

determined from time-lapse movie images using ImageJ software; n$58 cells from three independent experiments (right). Values are means 6 s.d., ***P,0.001

compared with control. (E) PC3 cells were transfected with RhoB-targeting siRNA (siRhoB1) or control siRNA (siControl). 24 hours later, cells were transfected

with a construct encoding GFP–RhoB or GFP alone. After 48 hours, cells were added to VN-coated wells. (Left) Representative phase-contrast images; scale bars:

50 mm. (Right) Spread area of cells after seeding was determined using Image J. Values are means 6 s.e.m., ***P,0.001 compared with control. (F) Cell speed

was determined by time-lapse microscopy, acquiring an image every 10 minutes over a period of 24 hours. Data are from 30 cells in two independent

experiments. Values are means 6 s.e.m., *P,0.01. (G) The volume and diameter of cells in suspension were measured. Graphs show data from three independent

experiments, each carried out in triplicate. Results are normalized to control. Values are means 6 s.d. (H) PC3 cells were transfected with uPAR-targeting siRNA

(si-uPAR1) or control siRNA (siControl). 24 hours later, cells were transfected with a construct encoding GFP–RhoB or GFP alone. After a further 10 hours, cells

were added to VN-coated wells and cultured in the absence of serum and imaged by time-lapse microscopy for 24 hours. After 24 hours, cell area was determined

from time-lapse movie images; n$100 cells pooled from three independent experiments. Values are means 6 s.d., *P,0.05 compared with siControl.

Journal of Cell Science 125 (10)2376

Journ

alof

Cell

Scie

nce

contrast, in metastatic prostate cancer cells RhoB promotesmigration (Yoneda et al., 2010), and RhoB depletion impairsTGF b-dependent migration (Vasilaki et al., 2010).

RhoB was required for uPAR-induced signal transduction totwo proteins involved in cell migration, cofilin and paxillin.Cofilin induces actin filament severing in lamellipodia and is

important for cell migration and invasion (Oser and Condeelis,2009). We observed rapid cofilin phosphorylation in response toATF, which is known to inhibit its activity. RhoA can stimulatecofilin phosphorylation through its downstream targets ROCK1

and ROCK2 (Oser and Condeelis, 2009). RhoB can also interactwith ROCKs in vitro and can act through ROCK1 to stimulateNFkB activation (Rodriguez et al., 2007), but whether it regulates

cofilin through ROCK is not known. uPA–uPAR was previouslyreported to increase tyrosine phosphorylation of paxillin as wellas other focal adhesion proteins in endothelial cells (Tang et al.,

1998). Because paxillin is activated by integrins and plays acentral role in coordinating integrin-based adhesion turnover(Deakin and Turner, 2008), it is likely that RhoB regulates

paxillin through its effects on integrins, and that paxillin thencontributes to uPAR-induced migration.

uPA–uPAR play an important role in cancer dissemination,

and thus downstream regulators of uPAR signalling could betargets for inhibiting cancer progression. Although RhoB isconsidered a tumour suppressor in some cancer cell types (Vega

and Ridley, 2008), our data indicate that it could also play a rolein cancer invasion as a downstream mediator of uPAR responses.

Materials and MethodsReagents

Polyclonal anti-uPA antibody was a gift from P. A. Andreasen, Aarhus, Denmark.Monoclonal anti-uPAR R4 antibody was kindly provided by G. Hoyer-Hansen,Finsen Institute, Copenhagen, Denmark. Polyclonal anti-uPAR 399 antibody andATF protein were purchased from American Diagnostica (Greenwich, CT).Polyclonal anti-RhoB and polyclonal anti-integrin b5 antibodies were from SantaCruz Biotechnology (Insight Biotechnology, Wembley, UK). Antibodies to RhoA(clone 26C4), RhoB (119), RhoC (C-16) and Cdc42 (clone B-8) were from SantaCruz Biotechnology, Rac1 (clone 23A) and anti-integrin av were from Millipore(Watford, UK), mouse anti-RhoJ, rabbit anti-RhoQ, goat anti-RhoU, mouse

anti-integrin b1 (clone P5D2) and mouse anti-active integrin b1 (clone 12G10)were from Abcam (Cambridge, UK). Monoclonal anti-RhoE antibody wasdescribed previously (Riento et al., 2003). Sheep anti-RhoF antibody was kindlyprovided by Harry Mellor, Bristol University, UK. Mouse anti-RhoG was a giftfrom Martin A. Schwartz (University of Virginia, USA). Polyclonal anti-phosphorylated Tyr118 paxillin was from Biosource, Invitrogen (Paisley, UK);monoclonal anti-paxillin (clone 349) was from BD Transduction Laboratories(Oxford, UK); polyclonal anti-phosphorylated Thr308 AKT was from CellSignaling (New England Biolabs, Hitchin, UK). Collagen type IV, polyclonalanti-tubulin and monoclonal anti-talin antibodies were from Sigma-Aldrich(Dorset, UK). Vitronectin, Transwell filters and Matrigel-coated Transwellfilters with a PET membrane with 8 mm pores were from BD Biosciences. ECLdetection kit was from Amersham Biosciences. The complete mini EDTA-freeprotease inhibitor was from Roche (Welwyn Garden City, UK). Halt phosphataseinhibitor cocktail was from Pierce Biotechnology (Rockford, USA). All cellculture reagents, Lipofectamine 2000 and Oligofectamine were purchased fromInvitrogen.

Cell culture and transfectionsPC3 prostate cancer cells were cultured in Roswell Park Memorial Institutemedium (RPMI) containing 10% FBS, L-glutamine (300 mg/ml), penicillin(100 IU/ml) and streptomycin (100 mg/ml). Cells were seeded at 36105 per wellin six-well plates and transfected with siRNAs (final concentration 50 nM) inantibiotic-free medium using Oligofectamine according to the manufacturer’sinstructions. All Rho GTPase siRNAs were obtained from Dharmacon (FisherScientific UK, Loughborough, UK). A non-targeting siRNA was used as a controlin all experiments (ON-TARGET control; Dharmacon). uPA and uPAR siRNAswere either from Qiagen or Sigma. All siRNA sequences are shown insupplementary material Table S1.

ImmunoblottingCells were harvested in lysis buffer (50 mM Tris-HCl pH 7.6, 2 mMEDTA, 150 mM NaCl, 0.5% Triton X-100) supplemented with 1 mMphenylmethylsulphonyl fluoride and 16 Complete mini EDTA-free proteaseinhibitor and PhosSTOP phosphatase inhibitor (Roche). Debris was removed bycentrifugation at 10,000 g for 20 minutes at 4 C, and protein content was assessedby a Bradford protein assay. 50 mg proteins per sample were separated by SDS-PAGE and transferred to Immobilon-P PVDF membranes (Millipore). Membraneswere subsequently incubated for 2 hours at room temperature in TBST buffer[125 mM Tris-HCl (pH 8.0), 625 mM NaCl, 0.1% Tween 20] containing 5% BSAand further incubated at 4 C for 16 hours with primary antibodies (used at adilution of 1:500). Secondary HRP-conjugated mouse (GE Healthcare, Chalfont StGiles, UK) or rabbit (Dako, Ely, UK) antibodies were used at a dilution of 1:5000.Membranes were developed using the enhanced chemiluminescence system (GEHealthcare). X-ray films were scanned and analysed using a Bio-Rad GS-800densitometer.

Fig. 8. RhoB depletion inhibits integrin association with uPAR.

(A) PC3 cells were lysed 48 hours after transfection with siRhoB2 or

siControl and analysed by immunoblotting with polyclonal anti-RhoB

antibody, or anti-tubulin antibodies as a loading control. Total lysates

were immunoprecipitated with anti-b1 or anti-av antibodies. Both total

lysates (inputs) and agarose bead eluates (IP) were subjected to

electrophoresis and subsequently probed with the same antibodies used

for immunoprecipitation, with R2 monoclonal anti-uPAR antibody and

monoclonal anti-talin antibody. (B) PC3 cells were transfected with

siRNAs, then serum-starved for 24 hours, and stained for surface levels

of active b1 integrin (12G10 antibody), then analysed by flow

cytometry. Values are the mean fluorescence of the population 6 s.d. of

two separate experiments each performed in duplicate, normalized to

total b1 integrin levels.

RhoB regulates uPAR signalling 2377

Journ

alof

Cell

Scie

nce

Co-immunoprecipitation

Immunoprecipitations were performed as previously described (Wei et al., 1996).

PC3 cells (5606/sample) were washed twice with microtubule stabilization buffer

(0.1 M PIPES, pH 6.9, 2 M glycerol, 1 mM EGTA, 1 mM magnesium acetate)

and then extracted in 0.2% Triton X-100 with additional protease inhibitors. The

insoluble residue, enriched in cytoskeleton-associated proteins, was solubilized in

RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 1% deoxycholate, 0.1%

SDS, 1% Triton-X-100, and protease inhibitors) and preincubated with non-

immune serum and 10% protein-A–Sepharose for 2 hours at 4 C. Aftercentrifugation, the supernatants were incubated with 2 ml of antisera to b1

integrin and av integrin, or with 2 ml of non-immune serum for 2 hours at 4 C.

After 30 minutes of incubation with 10% protein-A–Sepharose at room

temperature, the immunoprecipitates were washed, subjected to 9% SDS-PAGE

and analysed by western blotting using R2 monoclonal anti-uPAR antibody or

monoclonal anti-talin antibody at a concentration of 1 mg/ml. Membranes were

reprobed with monoclonal anti-b1 integrin or av integrin antibody.

Measurement of uPA secretion

Cells were seeded in 24-well plates and, 24 hours later, were serum-starved for

16 hours. Equal amounts of protein from the conditioned medium (20 mg/sample)

were concentrated by trichloroacetic acid precipitation, loaded under non-reducing

conditions and resolved by SDS-PAGE, then subjected to immunoblotting analysisusing a polyclonal anti-uPA antibody.

RhoB activity assay

The pull-down assay to measure RhoB activity was performed using a Rhoactivation assay kit, according to the manufacturer’s protocol (Cytoskeleton,

Denver, CO). Sub-confluent PC3 cells were serum-starved for 24 hours, then

treated with acid buffer (50 mM glycine, 100 mM NaCl, pH 3) for 2 minutes to

remove membrane-bound uPA. Cells were washed in serum-free medium. Cells

were stimulated in serum-free medium with 10 nM ATF for 5 minutes at 37 C

before addition of lysis buffer (l50 mM Tris-HCl pH 7.5, 1 mM EDTA, 500 mM

NaCl, 10 mM MgCl2, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS,

10% glycerol, 0.5% 2-mercaptoethanol) at 4 C. A protein assay was performed

(Bio-Rad) and equal amounts of total protein were used for each pull-down assay.Lysates were incubated with GST–rhotekin–RBD on glutathione–agarose beads

(Cytoskeleton) for 1 hour at 4 C, and then the beads were washed four times in

lysis buffer. The agarose beads were boiled in SDS–PAGE sample buffer to

release active Rho proteins, which were then processed for immunoblotting with

an anti-RhoB antibody. Total cell lysate (30 mg) per sample was used to detect

total RhoB.

Migration and invasion assays

Transwell filters coated with collagen IV (50 mg/ml, diluted in PBS) with a PET

membrane with 8 mm pores (BD Biosciences) were rehydrated for 2 hours at 37 C

in medium without supplements. Transfected cells were washed once in PBS, and

105 cells were seeded into the upper chamber of the Transwells in serum-free

medium containing 0.1% BSA. Medium containing 10 nM ATF and 0.1% BSAwas added to the bottom chamber as a chemoattractant. Control wells without ATF

were included to assess random migration. Cells were allowed to migrate for

4 hours at 37 C, in 5% CO2. For invasion assays the Transwell filters were coated

in Matrigel, and ATF or FCS was used as a chemoattractant, as indicated. The cells

were allowed to invade for 22 hours at 37 C and then the remaining cells and

matrix were removed from the upper side of the membrane. The cells on the

bottom part of the membrane were fixed in methanol containing 0.1% Crystal

Violet. Ten separate bright-field images were randomly acquired of each Transwell

filter using a Nikon TE2000-E microscope with a Plan Fluor 106objective (Nikon,Kingston, UK). The cells in each image were counted and analysed in comparison

with control-transfected cells. Migration or invasion in the absence of ATF

(random migration) was taken as 100% for each siRNA treatment, to control for

variations in final cell number.

For time-lapse microscopy, PC3 cells were seeded 24 hours after transfection in24-well plates at sub-confluency (random migration assays) or confluency (scratch

wound assays) in growth medium. For scratch wound assays, the confluent

monolayer was wounded after 24 hours by scraping with a pipette tip and then

imaged by time-lapse microscopy. During acquisition of movies, cells were

incubated in a humidified chamber at 37 C, 5% CO2 in serum-free medium. In

some experiments CO2 was not used and instead 25 mM HEPES was added. A

phase-contrast image was acquired every 10 minutes for 24 hours on a fully

motorized Nikon TE2000-E microscope with a Plan Fluor 106 objective usingMetamorph 5.01 software (Molecular Devices, Wokingham, UK). For random

migration assays, cell migration speed was quantified with ImageJ software using

the plug-in ‘manual tracking’. In each experiment, 30 randomly chosen cells were

tracked and their mean migration speed was calculated. The area occupied by cells

in scratch wounds was determined from time-lapse movie images taken at different

time after wounding, using ImageJ analysis software (http://rsb.info.nih.gov/ij).

Adhesion assays

Flat-bottomed 96-well microtiter plates were coated with 10 mg/ml vitronectin,

collagen or 1% heat-denatured BSA in PBS (uncoated plastic) as a negative

control, and incubated overnight at 4 C. The plates were then blocked for 1 hour at

room temperature with 1% heat-denatured BSA in PBS. Cells were harvested

using trypsin and allowed to recover for 1 hour at 37 C, then washed three times in

PBS. 105 cells were plated in each coated well and incubated for 1 hour at 37 C.

Attached cells were fixed with 3% paraformaldehyde in PBS for 10 minutes and

then incubated with 2% methanol for 10 minutes. The cells were finally stainedwith 0.5% Crystal Violet in 20% methanol. The stain was eluted using 0.1 M

sodium citrate in 50% ethanol, pH 4.2, and the absorbance at 595 nm was

measured in a spectrophotometer.

Flow cytometry analysis

PC3 cells were harvested in PBS containing 4 mM EDTA, washed in PBS

containing Ca2+ and Mg2+, and then 56105 cells were incubated with 10 mg/ml of

polyclonal anti-uPAR or polyclonal anti-integrin av, monoclonal anti-integrin

b1 or polyclonal anti-integrin b5 antibodies for 1 hour at 4 C. Purified

immunoglobulin was used as a negative control. The cells were then washed

and incubated with a fluorescein isothiocyanate-labelled goat anti-rabbit or anti-

mouse IgG for 30 minutes at 4 C. For analysis of surface-active b1 integrin levels,cells were scraped into PBS, fixed in 4% PFA–PBS for 30 minutes on ice and

blocked in 5% BSA–PBS for 30 minutes on ice. Cells were then incubated with

10 mg/ml anti-integrin b1 (12G10) overnight at 4 C. Finally, the cells were washed

and analysed by flow cytometry using a FACScan (Becton Dickinson, San Jose,

CA).

Quantitative PCR

PC3 cells were transfected with siRNAs. After 72 hours, cells were washed with

PBS, then total RNA was isolated by acid-phenol extraction using TRIzol Reagent

(Invitrogen) according to the manufacturer’s instructions. 1 mg of total RNA was

reverse-transcribed using QuantiTect Reverse Transcription (Qiagen, Hilden,Germany); 1 ml of a 1:10 dilution of the reverse transcription reaction was

analysed by qPCR with a Bio-Rad IQ5 system, using IQ2SYBR Green Supermix

for qPCR kit. The mRNAs measured were normalized to the internal

glyceraldehyde-3-phosphate dehydrogenase mRNA. Primers, designed using

Primer3 software and used at 0.25 mM, were as follows: for uPAR amplification,

forward primer 59-AAGGATACAGCTGGAGTCAG-39 and reverse primer 59-

GAGTTCATTCACTACCTGTTC-39; for RhoB amplification: forward primer 59-

CATTCTGACCACACTTGTACGC-39 and reverse primer 59- GGTTTCTTTTCC-

CTCTCCTTGT-39; for glyceraldehyde-3-phosphate dehydrogenase amplification,forward primer 59-ACATGTTCCAATATGATTCCA-39 and reverse primer 59-

TGGACTCCACGACGTACTCAG-39. The relative level of expression was

calculated with the formula 2–DDCt.

Statistical analysis

Statistical analysis was carried out where indicated using data from three or more

independent experiments each in triplicate, unless stated otherwise. Differences

between data sets were determined by using an unpaired Student’s t-test.

AcknowledgementsWe are grateful to Ritu Garg for technical support, Dario Gallottaand Roberta Cotugno for assistance with flow cytometry, FranciscoVega, Ferran Valderrama, Sarah Heasman and Philippe Riou forcomments and discussions, P. A. Andreasen, G. Hoyer-Hansen, H.Mellor and M. A. Schwartz for antibodies, and Nicolai Sidenius forproviding the construct encoding uPAR–GFP.

FundingThis work was supported by grants from Cancer Research UK [grantnumber C6620/A8833] to A.J.R., Italian Ministero dell’Istruzione,dell’Universita e della Ricerca (PRIN 2007) [grant number2007YJAP2M] to P.R.; Associazione Italiana per la Ricerca sulCancro [grant number IG 4714] to P.R.; the European MolecularBiology Organization [grant number ASTF 270.00-2007 to D.A.]; anda short-term fellowship from the Federation of the Societies ofBiochemistry and Molecular Biology to D.A.

Supplementary material available online at

http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.091579/-/DC1

Journal of Cell Science 125 (10)2378

Journ

alof

Cell

Scie

nce

ReferencesAguirre-Ghiso, J. A., Estrada, Y., Liu, D. and Ossowski, L. (2003). ERK(MAPK)

activity as a determinant of tumor growth and dormancy; regulation by p38(SAPK).

Cancer Res. 63, 1684-1695.

Alfano, D., Franco, P., Vocca, I., Gambi, N., Pisa, V., Mancini, A., Caputi, M.,

Vincenza Carriero, M., Iaccarino, I. and Stoppelli, M. P. (2005). The urokinase

plasminogen activator and its receptor: role in cell growth and apoptosis. Thromb.

Haemost. 93, 205-211.

Alfano, D., Iaccarino, I. and Stoppelli, M. P. (2006). Urokinase signaling through its

receptor protects against anoikis by increasing BCL-xL expression levels. J. Biol.

Chem. 281, 17758-17767.

Almasi, C. E., Brasso, K., Iversen, P., Pappot, H., Høyer-Hansen, G., Danø, K. and

Christensen, I. J. (2011). Prognostic and predictive value of intact and cleaved forms

of the urokinase plasminogen activator receptor in metastatic prostate cancer.

Prostate 71, 899-907.

Bai, S. W., Herrera-Abreu, M. T., Rohn, J. L., Racine, V., Tajadura, V.,

Survavanshi, N., Bechtel, S., Wiemann, S., Baum, B. and Ridley, A. J. (2011).

Identification and characterization of a set of conserved and new regulators of

cytoskeletal organization, cell morphology and migration. BMC Biol. 9, 54.

Bamburg, J. R. (1999). Proteins of the ADF/cofilin family: essential regulators of actin

dynamics. Annu. Rev. Cell Dev. Biol. 15, 185-230.

Blasi, F. and Carmeliet, P. (2002). uPAR: a versatile signalling orchestrator. Nat. Rev.

Mol. Cell Biol. 3, 932-943.

Blasi, F. and Sidenius, N. (2010). The urokinase receptor: focused cell surface

proteolysis, cell adhesion and signaling. FEBS Lett. 584, 1923-1930.

Calderwood, D. A. (2004). Integrin activation. J. Cell Sci. 117, 657-666.

Canguilhem, B., Pradines, A., Baudoin, C., Boby, C., Lajoie-Mazenc, I., Charveron, M.

and Favre, G. (2005). RhoB protects human keratinocytes from UVB-induced apoptosis

through epidermal growth factor receptor signaling. J. Biol. Chem. 280, 43257-43263.

Caswell, P. T., Vadrevu, S. and Norman, J. C. (2009). Integrins: masters and slaves of

endocytic transport. Nat. Rev. Mol. Cell Biol. 10, 843-853.

Chandrasekar, N., Mohanam, S., Gujrati, M., Olivero, W. C., Dinh, D. H. and Rao,

J. S. (2003). Downregulation of uPA inhibits migration and PI3k/Akt signaling in

glioblastoma cells. Oncogene 22, 392-400.

Connolly, E. C., Van Doorslaer, K., Rogler, L. E. and Rogler, C. E. (2010).

Overexpression of miR-21 promotes an in vitro metastatic phenotype by targeting the

tumor suppressor RHOB. Mol. Cancer Res. 8, 691-700.

Bousquet, E., Mazieres, J., Privat, M., Rizzati, V., Casanova, A., Ledoux, A., Mery,

E., Couderc, B., Favre, G. and Pradines, A. (2009). Loss of RhoB expression

promotes migration and invasion of human bronchial cells via activation of AKT1.

Cancer Res. 69, 6092-6099.

Danø, K., Behrendt, N., Høyer-Hansen, G., Johnsen, M., Lund, L. R., Ploug, M. and

Rømer, J. (2005). Plasminogen activation and cancer. Thromb. Haemost. 93, 676-

681.

Deakin, N. O. and Turner, C. E. (2008). Paxillin comes of age. J. Cell Sci. 121, 2435-

2444.

Dedhar, S., Saulnier, R., Nagle, R. and Overall, C. M. (1993). Specific alterations in

the expression of a3b1 and a6b4 integrins in highly invasive and metastatic variants

of human prostate carcinoma cells selected by in vitro invasion through reconstituted

basement membrane. Clin. Exp. Metastasis 11, 391-400.

Deng, G., Curriden, S. A., Wang, S., Rosenberg, S. and Loskutoff, D. J. (1996). Is

plasminogen activator inhibitor-1 the molecular switch that governs urokinase

receptor-mediated cell adhesion and release? J. Cell Biol. 134, 1563-1571.

Franco, P., Vocca, I., Vincenza Carriero, M., Alfano, D., Cito, L., Longanesi-

Cattani, I., Grieco, P., Ossowski, L. and Stoppelli, M. P. (2006). Activation of

urokinase receptor by a novel interaction between the connecting peptide region of

urokinase and avb5 integrin. J. Cell Sci. 119, 3424-3434.

Fritz, G., Kaina, B. and Aktories, K. (1995). The ras-related small GTP-binding

protein RhoB is immediate-early inducible by DNA damaging treatments. J. Biol.

Chem. 270, 25172-25177.

Harbeck, N., Kates, R. E., Look, M. P., Meijer-Van Gelder, M. E., Klijn, J. G.,

Kruger, A., Kiechle, M., Janicke, F., Schmitt, M. and Foekens, J. A. (2002).

Enhanced benefit from adjuvant chemotherapy in breast cancer patients classified

high-risk according to urokinase-type plasminogen activator (uPA) and plasminogen

activator inhibitor type 1 (n 5 3424). Cancer Res. 62, 4617-4622.

Heasman, S. J. and Ridley, A. J. (2008). Mammalian Rho GTPases: new insights into

their functions from in vivo studies. Nat. Rev. Mol. Cell Biol. 9, 690-701.

Huang, M. and Prendergast, G. C. (2006). RhoB in cancer suppression. Histol.

Histopathol. 21, 213-218.

Huang, M., Duhadaway, J. B., Prendergast, G. C. and Laury-Kleintop, L. D. (2007).

RhoB regulates PDGFR-beta trafficking and signaling in vascular smooth muscle

cells. Arterioscler. Thromb. Vasc. Biol. 27, 2597-2605.

Irigoyen, J. P., Munoz-Canoves, P., Montero, L., Koziczak, M. and Nagamine, Y.

(1999). The plasminogen activator system: biology and regulation. Cell. Mol. Life Sci.

56, 104-132.

Jahner, D. and Hunter, T. (1991). The ras-related gene rhoB is an immediate-early

gene inducible by v-Fps, epidermal growth factor, and platelet-derived growth factor

in rat fibroblasts. Mol. Cell. Biol. 11, 3682-3690.

Jo, M., Thomas, K. S., Somlyo, A. V., Somlyo, A. P. and Gonias, S. L. (2002).

Cooperativity between the Ras-ERK and Rho-Rho kinase pathways in urokinase-type

plasminogen activator-stimulated cell migration. J. Biol. Chem. 277, 12479-12485.

Kiian, I., Tkachuk, N., Haller, H. and Dumler, I. (2003). Urokinase-induced migrationof human vascular smooth muscle cells requires coupling of the small GTPases RhoAand Rac1 to the Tyk2/PI3-K signalling pathway. Thromb. Haemost. 89, 904-914.

Kjoller, L. and Hall, A. (2001). Rac mediates cytoskeletal rearrangements andincreased cell motility induced by urokinase-type plasminogen activator receptorbinding to vitronectin. J. Cell Biol. 152, 1145-2257.

Lebowitz, P. F., Davide, J. P. and Prendergast, G. C. (1995). Evidence thatfarnesyltransferase inhibitors suppress Ras transformation by interfering with Rhoactivity. Mol. Cell. Biol. 15, 6613-6622.

Liu, A. X., Rane, N., Liu, J. P. and Prendergast, G. C. (2001). RhoB is dispensable formouse development, but it modifies susceptibility to tumor formation as well as celladhesion and growth factor signaling in transformed cells. Mol. Cell. Biol. 21, 6906-6912.

Llinas, P., Le Du, M. H., Gardsvoll, H., Danø, K., Ploug, M., Gilquin, B., Stura,

E. A. and Menez, A. (2005). Crystal structure of the human urokinase plasminogenactivator receptor bound to an antagonist peptide. EMBO J. 24, 1655-1663.

Madsen, C. D. and Sidenius, N. (2008). The interaction between urokinase receptor andvitronectin in cell adhesion and signalling. Eur. J. Cell Biol. 87, 617-629.

Madsen, C. D., Ferraris, G. M., Andolfo, A., Cunningham, O. and Sidenius, N.(2007). uPAR-induced cell adhesion and migration: vitronectin provides the key. J.

Cell Biol. 177, 927-939.

Margheri, F., Manetti, M., Serratı, S., Nosi, D., Pucci, M., Matucci-Cerinic, M.,Kahaleh, B., Bazzichi, L., Fibbi, G., Ibba-Manneschi, L. et al. (2006). Domain 1 ofthe urokinase-type plasminogen activator receptor is required for its morphologic andfunctional, beta2 integrin-mediated connection with actin cytoskeleton in humanmicrovascular endothelial cells: failure of association in systemic sclerosis endothelialcells. Arthritis Rheum. 54, 3926-3938.

Monferran, S., Skuli, N., Delmas, C., Favre, G., Bonnet, J., Cohen-Jonathan-Moyal,E. and Toulas, C. (2008). avb3 and avb5 integrins control glioma cell response toionising radiation through ILK and RhoB. Int. J. Cancer 123, 357-364.

Montuori, N., Visconte, V., Rossi, G. and Ragno, P. (2005). Soluble and cleaved formsof the urokinase-receptor: degradation products or active molecules? Thromb.

Haemost. 93, 192-198.

Montuori, N., Bifulco, K., Vincenza Carriero, M., La Penna, C., Visconte, V.,Alfano, D., Pesapane, A., Rossi, F. W., Salzano, S., Rossi, G. et al. (2011). Thecross-talk between the urokinase receptor and fMLP receptors regulates the activity ofthe CXCR4 chemokine receptor. Cell. Mol. Life Sci. 68, 2453-2467.

Muller, S. M., Okan, E. and Jones, P. (2000). Regulation of urokinase receptortranscription by Ras- and Rho-family GTPases. Biochem. Biophys. Res. Commun.

270, 892-898.

Oser, M. and Condeelis, J. (2009). The cofilin activity cycle in lamellipodia andinvadopodia. J. Cell. Biochem. 108, 1252-1262.

Pulukuri, S. M., Gondi, C. S., Lakka, S. S., Jutla, A., Estes, N., Gujrati, M. and Rao,J. S. (2005). RNA interference-directed knockdown of urokinase plasminogenactivator and urokinase plasminogen activator receptor inhibits prostate cancer cellinvasion, survival, and tumorigenicity in vivo. J. Biol. Chem. 280, 36529-36540.

Rabbani, S. A., Ateeq, B., Arakelian, A., Valentino, M. L., Shaw, D. E.,

Dauffenbach, L. M., Kerfoot, C. A. and Mazar, A. P. (2010). An anti-urokinaseplasminogen activator receptor antibody (ATN-658) blocks prostate cancer invasion,migration, growth, and experimental skeletal metastasis in vitro and in vivo.Neoplasia 12, 778-788.

Ridley, A. J. (2006). Rho GTPases and actin dynamics in membrane protrusions andvesicle trafficking. Trends Cell Biol. 16, 522-529.

Riento, K., Guasch, R. M., Garg, R., Jin, B. and Ridley, A. J. (2003). RhoE binds toROCK I and inhibits downstream signaling. Mol. Cell. Biol. 23, 4219-4229.

Rodriguez, P. L., Sahay, S., Olabisi, O. O. and Whitehead, I. P. (2007). ROCKI-mediated activation of NF-kappaB by RhoB. Cell. Signal. 19, 2361-2369.

Sandilands, E., Cans, C., Fincham, V. J., Brunton, V. G., Mellor, H., Prendergast,

G. C., Norman, J. C., Superti-Furga, G. and Frame, M. C. (2004). RhoB and actinpolymerization coordinate Src activation with endosome-mediated delivery to themembrane. Dev. Cell 7, 855-869.

Schaller, M. D. and Parsons, J. T. (1995). pp125FAK-dependent tyrosinephosphorylation of paxillin creates a high-affinity binding site for Crk. Mol. Cell.

Biol. 15, 2635-2645.

Sehgal, I., Foster, T. P. and Francis, J. (2006). Prostate cancer cells show elevatedurokinase receptor in a mouse model of metastasis. Cancer Cell Int. 6, 21.

Sheng, S. (2001). The urokinase-type plasminogen activator system in prostate cancermetastasis. Cancer Metastasis Rev. 20, 287-296.

Smith, H. W. and Marshall, C. J. (2010). Regulation of cell signalling by uPAR. Nat.

Rev. Mol. Cell Biol. 11, 23-36.

Smith, H. W., Marra, P. and Marshall, C. J. (2008). uPAR promotes formation of thep130Cas-Crk complex to activate Rac through DOCK180. J. Cell Biol. 182, 777-790.

Stoppelli, M. P., Corti, A., Soffientini, A., Cassani, G., Blasi, F. and Assoian, R. K.(1985). Differentiation-enhanced binding of the amino-terminal fragment of humanurokinase plasminogen activator to a specific receptor on U937 monocytes. Proc.

Natl. Acad. Sci. USA 82, 4939-4943.

Sturge, J., Hamelin, J. and Jones, G. E. (2002). N-WASP activation by a b1-integrin-dependent mechanism supports PI3K-independent chemotaxis stimulated by uroki-nase-type plasminogen activator. J. Cell Sci. 115, 699-711.

Tang, H., Kerins, D. M., Hao, Q., Inagami, T. and Vaughan, D. E. (1998). Theurokinase-type plasminogen activator receptor mediates tyrosine phosphorylation offocal adhesion proteins and activation of mitogen-activated protein kinase in culturedendothelial cells. J. Biol. Chem. 273, 18268-18272.

RhoB regulates uPAR signalling 2379

Journ

alof

Cell

Scie

nce

Vasilaki, E., Papadimitriou, E., Tajadura, V., Ridley, A. J., Stournaras, C. andKardassis, D. (2010). Transcriptional regulation of the small GTPase RhoB gene byTGFbeta-induced signaling pathways. FASEB J. 24, 891-905.

Vassalli, J. D., Baccino, D. and Belin, D. (1985). A cellular binding site for the Mr55,000 form of the human plasminogen activator, urokinase. J. Cell Biol. 100, 86-92.

Vega, F. M. and Ridley, A. J. (2008). Rho GTPases in cancer cell biology. FEBS Lett.

582, 2093-2101.Vega, F., Fruhwirth, G., Ng, T. and Ridley, A. J. (2011). RhoA and RhoC have

distinct roles in migration and invasion by acting through different targets. J. Cell

Biol. 193, 655-665.Vincenza Carriero, M. and Stoppelli, M. P. (2011). The urokinase-type plasminogen

activator and the generation of inhibitors of urokinase activity and signaling. Curr.

Pharm. Des. 17, 1944-1961.Vincenza Carriero, M., Franco, P., Vocca, I., Alfano, D., Longanesi-Cattani, I.,

Bifulco, K., Mancini, A., Caputi, M. and Stoppelli, M. P. (2009). Structure,function and antagonists of urokinase-type plasminogen activator. Front. Biosci. 14,3782-3794.

Wei, Y., Lukashev, M., Simon, D. I., Bodary, S. C., Rosenberg, S., Doyle, M. V. and

Chapman, H. A. (1996). Regulation of integrin function by the urokinase receptor.

Science 273, 1551-1555.

Wheeler, A. P. and Ridley, A. J. (2004). Why three Rho proteins? RhoA, RhoB, RhoC,

and cell motility. Exp. Cell Res. 301, 43-49.

Wheeler, A. P. and Ridley, A. J. (2007). RhoB affects macrophage adhesion, integrin

expression and migration. Exp. Cell Res. 313, 3505-3516.

Wherlock, M., Gampel, A., Futter, C. and Mellor, H. (2004). Farnesyltransferase

inhibitors disrupt EGF receptor traffic through modulation of the RhoB GTPase.

J. Cell Sci. 117, 3221-3231.

Yoneda, M., Hirokawa, Y. S., Ohashi, A., Uchida, K., Kami, D., Watanabe, M.,

Yokoi, T., Shiraishi, T. and Wakusawa, S. (2010). RhoB enhances migration

and MMP1 expression of prostate cancer DU145. Exp. Mol. Pathol. 88, 90-95.

Zhou, J., Zhu, Y., Zhang, G., Liu, N., Sun, L., Liu, M., Qiu, M., Luo, D., Tang, Q.,

Liao, Z. et al. (2011). A distinct role of RhoB in gastric cancer suppression. Int. J.

Cancer 128, 1057-1068.

Journal of Cell Science 125 (10)2380

Journ

alof

Cell

Scie

nce